Wearable sensor activity analysis using semi-Markov models with a grammar

被引：26

作者：

Thomas, O. ^{[1
]}

Sunehag, P. ^{[1
]}

Dror, G. ^{[2
]}

Yun, S. ^{[3
]}

Kim, S. ^{[3
]}

Robards, M. ^{[1
]}

Smola, A. ^{[4
]}

Green, D. ^{[5
]}

Saunders, P. ^{[5
]}

机构：

[1] NICTA, Canberra, ACT 2601, Australia

[2] Acad Coll Tel Aviv Yaffo, Sch Comp Sci, IL-61083 Tel Aviv, Israel

[3] Korea Adv Inst Sci & Technol, Taejon 305701, South Korea

[4] Yahoo Res, Santa Clara, CA 95050 USA

[5] Australian Inst Sport, Dept Physiol, Belconnen, ACT 2616, Australia

来源：

PERVASIVE AND MOBILE COMPUTING | 2010年 / 6卷 / 03期

关键词：

Activity recognition; Semi-Markov models; Machine learning; Human performance; Accelerometer; GAIT EVENT DETECTION; CHILD;

D O I：

10.1016/j.pmcj.2010.01.002

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

080201 [机械制造及其自动化];

摘要：

Detailed monitoring of training sessions of elite athletes is an important component of their training. In this paper we describe an application that performs a precise segmentation and labeling of swimming sessions. This allows a comprehensive breakdown of the training session, including lap times, detailed statistics of strokes, and turns. To this end we use semi-Markov models (SMM), a formalism for labeling and segmenting sequential data, trained in a max-margin setting. To reduce the computational complexity of the task and at the same time enforce sensible output, we introduce a grammar into the SMM framework. Using the trained model on test swimming sessions of different swimmers provides highly accurate segmentation as well as perfect labeling of individual segments. The results are significantly better than those achieved by discriminative hidden Markov models. (C) 2010 Published by Elsevier B.V.

引用

页码：342 / 350

页数：9

共 25 条

[1]

[Anonymous], 1994, 1994 1 WORKSH MOB CO, DOI DOI 10.1109/WMCSA.1994.16

[2]

Aziz O, 2006, BSN 2006: International Workshop on Wearable and Implantable Body Sensor Networks, Proceedings, P13

[3]

A MAXIMIZATION TECHNIQUE OCCURRING IN STATISTICAL ANALYSIS OF PROBABILISTIC FUNCTIONS OF MARKOV CHAINS [J].